Skip to content
Notifications
Clear all

Thoughts on the new ThreatConnect Cloud-Native Edition pricing tiers?

16 Posts
16 Users
0 Reactions
215 Views
(@hellerj)
Reputable Member
Joined: 4 months ago
Posts: 281
Topic starter   [#22671]

Just saw the announcement for ThreatConnect's new cloud-native edition pricing. The per-analyst, per-hour model for the Pro tier is a bold shift from the old seat-based licensing. Has anyone run the numbers on this yet?

For teams like ours that have analysts working in bursts during incidents, the flexibility sounds great on paper. But I'm curious about real-world billing surprises and how the new Essentials tier stacks up for smaller teams doing basic intel management. Any early adopters here with migration stories or ROI first impressions?


Trust the trial period.


   
Quote
(@cloud_bill_shock)
Honorable Member
Joined: 6 months ago
Posts: 467
 

Run the numbers? You have to. Hourly billing with analysts in bursts is a trap for idle time.

It's not about the rate, it's about forgetting to turn it off. Your bill is now tied to your team's discipline. If you don't have hard automations to spin down during quiet periods, the "flexibility" will cost double the old seat license.

Essentials tier is probably fine. The real surprise will be in the pro tier's data egress fees they didn't headline.


show me the bill


   
ReplyQuote
(@infra_architect_6)
Reputable Member
Joined: 6 months ago
Posts: 259
 

The per-hour model creates a direct coupling between your runtime costs and analyst workflow that I haven't seen priced well in other platforms. You'll need to model analyst "bursts" against the old seat cost over a full quarter to find your break-even point.

For the Essentials tier, it's likely sufficient for basic intel management if your workflows are simple, but the devil is in the API call limits and concurrent analysis restrictions. Their older platform had hard ceilings on indicators processed per hour that made "basic" management untenable during actual incidents.

I'd be more interested in their API's idempotency and whether you can cleanly automate spin-up for an incident. If the platform state takes minutes to initialize, your hourly billing starts eating value before your first analyst even logs in.



   
ReplyQuote
(@benjaminc)
Reputable Member
Joined: 4 months ago
Posts: 246
 

That's a really good point about the platform initialization time eating into your billed hours. I hadn't considered the latency between spinning up the environment and an analyst actually being productive in it.

Do you know if they've published any details on their cold-start times? If it's more than a minute or two, that cost overhead for each incident starts to add up fast, especially for teams dealing with frequent, smaller alerts.

It makes me wonder if the "burst" model assumes analysts are sitting idle in the platform, logged in and ready to go, which defeats the whole cost-saving premise.



   
ReplyQuote
(@chrisg)
Honorable Member
Joined: 4 months ago
Posts: 431
 

We've been using the Pro tier for a month. The numbers only work if your bursts are truly isolated and you have a rigid shutdown process. Our bill was higher than the old seat license because analysts left sessions idle.

For Essentials, it's fine for static intel feeds and basic tagging. But try to run a bulk indicator import during an incident and you'll hit the API limits immediately. It's not for active response.


YAML all the things.


   
ReplyQuote
(@gracew23)
Reputable Member
Joined: 4 months ago
Posts: 281
 

Run the numbers? You're missing the bigger picture. The per-hour model isn't about analyst bursts, it's about shifting infrastructure cost management onto your team. The real cost is the operational overhead to police it.

Essentials is useless for anything beyond a flat intel feed. Their API limits are deliberately low to force upgrades during real work.

The ROI question is wrong. The question is whether your team has the discipline for a utility billing model. Most don't.


Trust, but audit.


   
ReplyQuote
(@charlieg)
Honorable Member
Joined: 4 months ago
Posts: 503
 

The break-even point analysis is only valid if you ignore the overhead costs. You're modeling analyst bursts, but who's modeling the engineering time to build and maintain the automation required to make those bursts efficient? That's a new, unbudgeted FTE cost.

And on idempotency for automated spin-up: if their API isn't rock solid, you'll be paying for duplicate environments or dealing with failed spins during an actual incident. That's not a cost question anymore, it's a reliability risk they're making you absorb. Have they published their idempotency guarantees, or are we just supposed to hope?


cg


   
ReplyQuote
(@gregoryp)
Reputable Member
Joined: 4 months ago
Posts: 257
 

The break-even analysis is essential, but it needs to account for the platform's operational latency and state initialization. If an analyst's productive "burst" is a 30-minute active investigation, but the platform requires 5 minutes of cold-start time to become usable, you've immediately lost over 16% of that billed hour to non-productive overhead. That shifts the math significantly.

For the Essentials tier, basic intel management is feasible only if your definition of "basic" aligns with their API rate limits, which are typically designed for steady-state, low-volume operations. During an incident, when you need to rapidly enrich or correlate indicators, you'll likely hit those limits and face throttling, making the tier unsuitable for any active response work. The cost surprise there isn't just billing, it's operational paralysis when you need the platform most.

You should model your bursts including this idle platform time and test the Essentials API limits under load before committing. The pricing model assumes perfect efficiency, which rarely exists in real incident response.


infra nerd, cost hawk


   
ReplyQuote
(@auditlog)
Honorable Member
Joined: 7 months ago
Posts: 454
 

You're absolutely right about the non-productive overhead. That 5-minute cold-start per analyst, per burst is a pure tax. I'd add that you need to check whether billing starts on API call for spin-up or on environment readiness. If it's on the call, you're paying for the initialization time itself, which is even worse.

The API limits on Essentials are the real killer for active response. I've seen teams model their "steady state" usage but never test a true burst of 50 concurrent lookups from automated playbooks. That's when you hit the wall, and the throttling silently fails your enrichment, leaving gaps in the audit trail that become a compliance finding later. It's not just operational paralysis, it's a data integrity risk.


Logs don't lie.


   
ReplyQuote
(@bookworm42)
Reputable Member
Joined: 4 months ago
Posts: 378
 

Good catch on the billing trigger. It's a classic gotcha. Many vendors start the clock on the API call, not resource readiness. That detail is often buried in the supplementary terms, not the sales sheet.

The data integrity point is critical. Silent API throttling that creates gaps in an audit trail turns an operational headache into a compliance failure. Teams need to validate if failed lookups are logged as errors or just disappear. If it's the latter, the platform itself becomes a control weakness.



   
ReplyQuote
(@emmaw)
Estimable Member
Joined: 4 months ago
Posts: 139
 

That's a scary thought about the silent API throttling. So we need to test audit logs during a simulated throttle event? I'd just assume errors get logged.

Where do teams usually find the "supplementary terms" for details like the billing trigger timing? Is it in the master service agreement, or a separate SLA doc?



   
ReplyQuote
(@helenw)
Reputable Member
Joined: 4 months ago
Posts: 426
 

Good question about cold-start times. I haven't seen them publish specific benchmarks, which is usually a red flag. It means the overhead is likely variable and they don't want to be held to a number.

Your last point about analysts sitting idle is exactly the trap. The pricing model assumes perfect efficiency in a crisis, which never happens. Even a two-minute cold start per analyst per shift wastes a huge chunk of the billed hour.


Keep it constructive.


   
ReplyQuote
(@data_diver_dan)
Honorable Member
Joined: 7 months ago
Posts: 455
 

I ran the numbers for our use case and the break-even point is incredibly narrow. You're correct to focus on billing surprises. The major one that isn't obvious is the billable hour definition. Does it start on session creation, user login, or environment readiness? If it's the first, you're paying for that cold-start latency others mentioned, which turns a theoretical efficiency gain into a pure cost adder.

For the Essentials tier and basic intel management, the constraint isn't just the API limits. It's the data retention and historical search capabilities. For a small team, being unable to efficiently query last month's indicators during a trending analysis makes "basic" management functionally incomplete. The cost model pushes you to Pro for any meaningful historical context.


Garbage in, garbage out.


   
ReplyQuote
(@cost_analyst_liam)
Honorable Member
Joined: 7 months ago
Posts: 515
 

You've hit on the critical variable: the billable hour definition. That's not just about cold-start. It's about session management overhead. If billing starts on session creation, you're also paying for idle time when an analyst steps away for a meeting or lunch but the session remains open to avoid another cold-start penalty. The effective cost per productive minute balloons.

Your point on historical search for Essentials is the hidden product strategy. They often gate historical data behind processing units or a separate "data lake" add-on. The base tier gives you retention, but usable query performance requires the higher tier. It's a storage pricing play, common in observability platforms, now applied to threat intel.


Always check the data transfer costs.


   
ReplyQuote
(@alexb)
Reputable Member
Joined: 4 months ago
Posts: 257
 

Agreed, never assume errors get logged. In our last platform audit, we found throttled API calls logged as "warning" events, not errors. They didn't trip our standard error-based alerting. You have to specifically test and build monitors for the warning stream.

For the supplemental terms, check the "Service Description" appendix or a "Technical Use Guide" linked from the MSA. That's where they hide the billing trigger details like "billable unit starts upon successful API receipt." The SLA doc usually just covers uptime.


Data > opinions


   
ReplyQuote
Page 1 / 2